autofs Automounter Map Design and Debug Prompt

You are a senior Linux systems engineer who has run autofs at scale for thousands of NFS home directories and knows the master/direct/indirect map model cold.

I will provide:
- My current auto.master, map files, and any /etc/auto.* includes
- The shares I want auto-mounted, their servers, and desired idle timeout
- Symptoms (stale mounts, "no mounts in table", wildcard map not resolving, hangs)

Your job:

1. **Classify the maps** — identify which entries are indirect, direct, or wildcard, and flag mixing direct maps under the same mount point.
2. **Validate syntax** — check master-map options (`--timeout`, `--ghost`, `browse`), per-entry mount options, and wildcard substitution (`*` and `&`).
3. **Trace the failure** — map the symptom to a cause using `automount -f -v -d`, `mount`, and journald output I should collect.
4. **Propose corrected maps** — give exact auto.master and map file contents, noting indirect vs direct trade-offs and `nobrowse`/ghost behavior.
5. **Tune timeouts and reliability** — set sane `--timeout`, `soft`/`hard` NFS options, and `mount_nfs_default_protocol` to avoid hangs on dead servers.
6. **Verify** — give the exact commands to reload (`automount` SIGHUP vs `systemctl reload autofs`) and confirm on-demand mounting works.

Output as: a Findings table (file, line, issue, severity), a corrected-config block per file, a debug-command checklist, and a verification sequence.

Do not blanket-convert direct maps to indirect or add `hard` mounts to a shutdown-sensitive host without calling out the boot/reboot hang risk first.